The goal of this research is to understand how we see what we see: how does the brain process the light falling on the retina of the eye to produce our perception of a world full of objects, people, things and places? During the past year we have continued to investigate the interaction between bottom up (sensory driven) and top-down (internally driven) processing in the brain, focusing on working memory and effect of behavioral goals (task), using complex visual stimuli, such as visual objects and scenes. 1) Mental Imagery Mental imagery, which occurs in in the absence of sensory input, relies entirely on top-down signals. Previously, we conducted a detailed comparison of visual imagery and perception for individual complex objects using fMRI. We found that (1) we can decode the identity of the specific object participants view or imagine in multiple brain regions, and (2) imagery and perceptual information are distributed differently throughout the visual processing stream. This prior work focused on imagery of objects that participants had just seen, requiring short-term memory only. We are currently extending this work to investigate imagery of objects from long-term memory, training subjects to learn particular objects on one day and then imagine them on a later day while we measure brain responses. 2) Working memory Working memory refers to the process of actively holding information in mind over short time intervals. As with mental imagery, working memory occurs in the absence of sensory input. We investigated which brain regions are involved in maintaining information during working memory and how the involvement of brain regions varies according to the behavioral goals of the observer (Lee, Kravitz, Baker, 2013, Nature Neuroscience). Participants performed two tasks on identical visual input (pictures of objects) with one task requiring participants to maintain visual properties and the other non-visual properties (name of the object). We found that during the maintenance of visual properties, object identity could be decoded from brain activity in visual, but not prefrontal, cortex, whereas the opposite held for non-visual properties. Thus, the ability to maintain information during working memory is a general and flexible property of the brain, with the role of individual regions being goal-dependent. 3) Behavioral goals/task Different tasks require different types of information to be extracted from visual stimuli, depending on the behavioral goals of the observer. We have been investigating how the representations of complex visual stimuli vary according to the task a participant is performing. First, we found that we could decode the task a participants was performing on a given visual object from activity in multiple regions throughout the brain. Further, we found a strong distinction between tasks that emphasized physical properties of the visual stimuli (e.g. color) and tasks that emphasized conceptual properties (e.g. real-world size). Second, we are now extending this work to visual scenes. Previously, we found that scene representations in a region of the brain thought to be critical for scene recognition primarily reflect the spatial properties of scenes (e.g. whether they are open or closed) and not the semantic properties (i.e. scene category). In our current work, we are investigating how these representations change according to task, by asking participants to focus on particular aspects of the scenes presented. Elucidating how the brain enables us to recognize objects, scenes, faces and bodies provides important insights into the nature of our internal representations of the world around us. Understanding these representations is vital in trying to determine the underlying deficits in many mental health and neurological disorders.